fix(types): expose PyTorch dataset compatibility - #8457
Open
uuzzrm wants to merge 1 commit into
Open
Conversation
Make Dataset and IterableDataset visible as PyTorch datasets to static type checkers without importing torch at runtime. Add a regression test for the optional dependency boundary.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do? Make
DatasetandIterableDatasetvisible to static type checkers as compatible with PyTorch's dataset classes. The runtime behavior remains unchanged: -torchis only imported underTYPE_CHECKINGfor the new bases. - The existing runtime parent-class behavior forIterableDatasetis preserved. -Datasetis not changed into a runtime subclass oftorch.utils.data.Dataset. This addresses the typing gap described in #7500, so Pyright can accept both map-style and iterable datasets astorch.utils.data.DataLoaderinputs, includingdataset.with_format(torch). ## Testing -ruff check src/datasets/arrow_dataset.py src/datasets/iterable_dataset.py tests/test_arrow_dataset.py-ruff format --check src/datasets/arrow_dataset.py src/datasets/iterable_dataset.py tests/test_arrow_dataset.py-git diff --check- Focused import regression: passed - Pyright 1.1.411 with a minimal PyTorch type stub coveringDataset.__getitem__,IterableDataset, andDataLoader: 0 errors - RelatedDatasetandIterableDatasettests: 804 passed, 108 skipped; PyTorch-dependent cases could not run because PyTorch is not installed in the local environment Fixes #7500 The implementation was prepared with Codex assistance and reviewed against the repository contribution guidelines.AI assistance
AI assistance was used for repository research, implementation, and test drafting. The diff and validation results are documented here for maintainer review.